In the light of the ongoing debate about optimum eye-relief, this Dutch website has interesting test data which suggests that useable eye-relief figures are actually much less than manufacturers claim. For example 15mm for Swaro 8.5, 13 mm for 8x 42 Ultravid, 14 mm for Zeiss 8 x 32. Also, there appear to be discrepancies in the claimed weights for binoculars. No data for Nikon unfortunately.
Worth a look!
Dave Griffiths.
http://www.tvwg.nl/testrapporten/kijkers.htm
I second Dave's endorsement. Jan Meijerink knows his stuff and does his tests thoroughly. Also, since he has his own standard measuring protocol, his results between different models and brands are comparable. I believe, judging by the results he gets, that he uses a method which measures eye-relief from the plane provided by the uppermost edge of the eyecup in the fully in position. This is what I do as well, and it means that depending on the curvature of the eyelens, there is a smaller or greater discrepancy between the figure thus obtained and one that would be measured from the surface of the eyelens at the dead center. This could also explain the difference between our measured results and the manufacturers' specs, but I feel that this "effective eye-relief" is more meaningful.
Another very useful bit of information Jan Meijerink provides is his freezer tests for internal condensation. He usually explains the results but also rates it on a scale from 1 to 5.
Very good pages, worth struggling a bit to understand the Dutch.
Kimmo